Lab Seminars

Talks, tutorials, and reading-group presentations from my lab seminars.

OptiML Lab Group Meetings

10 talks
  1. What I’ve been up to nowadays + What I’ll be doing in the future
  2. Bandits 101: A Maximally Non-technical Tutorial
  3. Improved Regret Bounds of (Multinomial) Logistic Bandits via Regret-to-Confidence-Set Conversion
  4. Introduction to Reinforcement Learning with Human Feedback (RLHF): A Theoretically Biased Overview
  5. Community Detection in Block Models: From SBMs to Block Markov Chains
  6. An Informal, Personal Presentation: My Experience at AISTATS 2023 & My Future Research Plan
  7. A Primer on (Combinatorial) Bandits
  8. Exact Dyanmics of Stochastic Gradient Descent in High Dimensions and Volterra (Integral) Equations
  9. Nearly Optimal Latent State Decoding in Block MDPs
  10. Fast and Efficient MMD-based Fair PCA via Optimization over Stiefel Manifold

OSI Lab Seminars

22 talks
  1. Bandits 101: A Maximally Non-technical Tutorial
  2. Flooding with Absorption: An Efficient Protocol for Heterogeneous Bandits over Complex Networks
  3. Conference Day - Theory Division
  4. Improved Regret Bounds of (Multinomial) Logistic Bandits via Regret-to-Confidence-Set Conversion
  5. Introduction to Reinforcement Learning with Human Feedback (RLHF): A Theoretically Biased Overview
  6. Community Detection in Block Models: From SBMs to Block Markov Chains
  7. A Primer on (Combinatorial Semi-) Bandits
  8. Fair Streaming Principal Component Analysis: Statistical and Algorithmic Viewpoint
  9. Improved Sample Complexity for Reward-free Reinforcement Learning under Low-rank MDPs
  10. Exact Dynamics of Stochastic Gradient Descent in High Dimensions and Volterra (Integral) Equations
  11. Entropic variants of SGD
  12. Conference Week (AISATS & ICML 2022)
  13. From Generalized Linear Bandit to Logistic Bandit: An Overview
  14. Gradient Descent on Infinitely Wide Neural Networks: Global Convergence and Generalization
  15. Clustering in Block Markov Chains
  16. Fast and Efficient MMD-based Fair PCA via Optimization over Stiefel Manifold
  17. Poisson Learning: Graph Based Semi-Supervised Learning At Very Low Label Rates
  18. Landscape and training regimes in deep learning
  19. How Powerful are Graph Neural Networks?
  20. Conference Week (NeurIPS 2020)
  21. Heavy-tail behaviour of SGD - Part 2
  22. Heavy-tail behaviour of SGD - Part 1

MDLT+P Seminar

4 talks
  1. Gradient Descent on Infinitely Wide Neural Networks
  2. Landscape and training regimes in deep learning
  3. Entropic variants of SGD
  4. Continuous Heavy-Tailed Theory of SGD

IBS BIMAG Journal Clubs

2 talks
  1. Introduction to Bayesian ML/DL, with Application to Parameter Inference of Coupled Non-linear ODEs - Part 2
  2. Introduction to Bayesian ML/DL, with Application to Parameter Inference of Coupled Non-linear ODEs - Part 1